Two experiments examined imperfect automation decision aids for maritime collision avoidance. In Experiment 1, the algorithm was driven purely by safety, recommending turning your ship in the direction that produced the greatest separation from a hazard ship. In Experiment 2, the algorithm incorporated two additional factors known to influence ship collision avoidance maneuvers: efficiency and procedural adherence to established “rules of the road.” In both experiments, results revealed heavy but not total reliance on the aid. A strong influence of categorical rules of the road was indicated by compliance with recommendations in concurrence to those conventions, even as they reduced continuous parameters of safety, particularly in Experiment 2 when the rules were incorporated in the algorithm. This illustrates the powerful influence of categorical procedural algorithm elements over continuous quantitative ones in affecting automation compliance. Results also revealed the dissociation between rated trust in, and behavioral dependence on decision aiding automation.

Exploration and exploitation are commonly cited in search and rescue scenarios to explain the process by which individuals work in a team and gather information about their environment (exploration) and identify potential solutions and adaptations (exploitation) to pursue successful outcomes. In this paper, we discuss exploration and exploitation as critical design features and highlight the importance of balancing them when designing team-based search and rescue missions. To test the proposed design decisions, we developed a usability study that includes two missions wherein teams consisting of three participants are tasked to rescue victims within a Minecraft-based 3D testbed.

This study investigated in 22 teams, individual and team trust measures reported by two human participants, recruited from a university populace, as they interacted with each other, robotic combat vehicles (RCVs), a human superior, and their team during a simulated Next Generation Combat Vehicle (NGCV) mission conducted within Minecraft. Trust was measured via survey questions based on established metrics and was found to be high toward the human peer, the human superior, and the overall team throughout the mission. In contrast, overall trust in the RCV was significantly lower in phases of the mission when breakdown in RCV functionality caused a hindrance in mission completion. Trust in the RCV was shown to recover as the mission progressed in phases without RCV maintenance issues. The findings reinforce that trust is distinguishable at the individual level and not necessarily perceived the same at the team level.